Linear monitor
Health-check Linear projects for slipping dates, stale In Progress, and orphan issues. Use when the user asks for a Linear health check, atRisk scan, monitoring, /linear-monitor, or when a Cursor Automation runs the Linear tracking health recipe. Default read + comment; no mass creates. NOT for bootstrap (linear-setup) or full rescue (linear-sync).From its SKILL.md
npx -y skills add tjcages/linear --skill linear-monitorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.6 KB, as published. Nobody here has run it
Linear monitor
Chat rules: RESPONSE.md. Auth: AUTH.md. Cadence: METHODOLOGY.md §6, §9, §20. Automation recipe: AUTOMATION.md.
Step 0 — auth
Probe Linear MCP. On failure → AUTH.md; stop.
Scope
User-named team/project, or all accessible teams if unset. Never soft-nudge git repos from Automation runs.
Checks (do now)
- Dates — milestones/issues with targets approaching where work is not trending Done → candidates for
atRisk(post only if health actually changes). - Stale In Progress — no comment >7 days → list with links.
- Orphans — issues with no milestone → list.
- Anti-patterns glance — milestone spam, status theater (§19); note only, don’t rewrite unasked.
Writes
- Default: read + optional comment on existing issues with findings.
- Status updates: only on real health changes (§20).
- Never: mass-create issues; never reorganize structure (hand off to linear-sync).
Output shape
- Findings first (≤5 bullets; overflow → later).
- What you changed (if anything), with links.
- ONE next action for the human.